Searching for the neurite density with diffusion MRI: Challenges for biophysical modeling

Area of Science:

• Neuroimaging
• Biophysics
• Diffusion MRI

Background:

• In vivo neurite density mapping with diffusion MRI (dMRI) faces challenges.
• Uncertainty exists regarding neurite diffusion properties and confounding factors like T2 relaxation.
• The validity of parameters as neurite density indices requires further exploration.

Purpose of the Study:

• Investigate challenges in neurite density estimation using dMRI.
• Assess the role of b-tensor encoding and multiple echo times.
• Explore the validity of neurite density indices in healthy and lesioned brain tissue.

Main Methods:

• Acquired data using b-tensor encoding and multiple echo times.
• Examined brain regions with low orientation coherence and white matter lesions.
• Utilized biophysical models to analyze diffusion properties.

Main Results:

• Microscopic anisotropy from b-tensor data correlates with myelinated axons, not dendrites.
• Unbiased density estimates in lesions need compartment-specific T2 values.
• 'Stick' fractions from models are not universally reliable neurite density indices.

Conclusions:

• Neurite density estimation necessitates accounting for differing axonal and dendritic diffusion and T2 properties.
• Constrained index parameters may be valid only within specific, yet-to-be-defined domains.
• Future studies must delineate the precise domains of validity for these indices.